unterhausen

Semi-autonomous BMW will ?fight driver? to deliver close passes of cyclists | road.cc

when I first saw this story, I thought the problem was that it didn't detect cyclists. But now I realize it's just worrying about going over the lane line more than it worries about passing distance. My toyota would brake in this situation. Of course, it will not do lane following at 25mph or less, so it might not come up.

Ninety5rpm

These Level 2 systems are awful. I wish everyone would just to do what Waymo is doing: Level 5 or manual.

caloso

So, identical to the majority of human drivers who consistently display more care for paint than fellow road users?

kevindsingleton 

Death? In a close pass? How did the cyclist die? From fear? From pancreatic cancer? Old age?

CB HI

Sadly, some think that a cyclist hit by a side mirror is nothing more than a close pass.

kevindsingleton 

Who got hit by a mirror? Did we enter the Twilight Zone, and we're now in a completely different thread? The topic is "Semi-autonomous BMW forces close pass of cyclists", not "Semi-autonomous BMW smacks cyclists with side mirror, then giggles to itself". (Emphasis mine.)

The article says, "Honest John writes: “If the road ahead is clear apart from a solitary cyclist, you do need to signal to overtake him, otherwise the steering wheel will fight you and you could pass him uncomfortably close.”"


So, the solution is to use the turn signal, which, as we all know, is an optional accessory on BMW motor vehicles, and is never, ever selected by the owners, which is the real story, here.

salcedo

This sounds like BMW forgot to consider cyclists when designing or "training" its semi-autonomous software. This is a big omission in a country like the UK with so many car-bicycle interactions

salcedo

Are you implying that close passes are not dangerous and cannot lead to accidents and death?

jimincalif

This! Well played!

CB HI

Sounds like he demands a dead cyclist before he considers it a problem.

Cyclist0108

I've had this kind of thing happen with my new Toyota Rav4. The "semi-autonomous" correction is fairly subtle, but it irks me that it does this at all. There is an over-ride button, (and a turn signal, which on the open road is fine, but around blind corners there might not be enough time), but I would rather leave the lane change detection on, and have the other settings on standard. Mine cuts out under ~35mph, and on the windy narrow mountainous roads where I live, it is definitely an issue.

Cyclist0108

In California, the use of turn signals is considered by many to be giving away information to the enemy.

salcedo

I'm never driven in California. But I have driven in places where people don't signal because it makes it harder to switch lanes.

That is a symptom of a bad driving culture.

Ninety5rpm

Worrying about human drivers not signaling now, at the end of the human-driven-automobile era, is like worrying about an egg shortage on the Titanic.

salcedo

That's a fair point.

I might be wrong, but I think we still have a very long time before self-driving cars take over worldwide.

https://www.youtube.com/watch?v=KnPiP9PkLAs

kevindsingleton 

You didn't answer any of my questions. Why should I answer yours?

kevindsingleton 

Certainly before I consider "who would be liable to death of a cyclists". Indeed, it's practically necessary.

JoeyBike

Or a sign of weakness.

I-Like-To-Bike

No habeas corpus necessary on the Soap Opera Drama Show available on A&S 24/7.

salcedo

I thought the questions that you asked were rhetorical. I didn't think you were expecting an answer. I will try to answer now.

I suspect that sometimes cyclists die due to accidents that could be avoided if cars didn't pass cyclists that closely. The cyclist in the story has not died yet as far as I know.

Nobody in the story linked was hit by a mirror. We are not in a completely different thread. It is the same thread. You can verify this by scrolling up.


It sounded to see like you were trying to make a point. But the point that you were trying to make is not very clear to me. If you did want to make a point, it would be better if you made it explicitly, by answering my question. But you don't have to if you don't want to.

I'll ask again in case you do what to make the point. Are you trying to imply that close passes are not dangerous and cannot result in death?

kevindsingleton 

You didn't read the article. Why are you arguing, if you didn't read the article? (Note: that question is not rhetorical). How closely is "that closely"? (Note: that question is not rhetorical, either).


You don't know, because you didn't read the article.


Thanks.

I'll concede that a close pass could be deemed a proximate cause of a crash that could result in death. However, since there was no close pass "in this case", and no cyclist that crashed "in this case", I still consider it a monumental leap to begin assigning liability "in this case".

salcedo

Jesus, you are a feisty one. Firs of all, your question is ill posted because I did read the article, and the articles linked inside the article, and I saw the video referenced in the article of a semi-autonomous car doing a close pass during a test drive. But I'll admit that my memory of the article is a bit fuzzy and I didn't pay too much attention when answering your rhetorical questions. Because what I wanted to do is get an answer.

Why am I arguing? Again, because I wanted to understand your point, which you still seem reluctant to make clearly. I don't have a strong opinion about how dangerous are the passes that people on YT label as close passes. That is why I am asking your opinion.

How closely is "that closely"? It depends largely on context. If a compact car passes me very closely at 12mph when I'm riding 10mph, I would feel a lot safer than if a semi doing 65 passed 6 ft away.

Again, your original post was "Death? In a close pass? How did the cyclist die? From fear? From pancreatic cancer? Old age?" But, I'll take it from your new answer that was not the point you were trying to make


Just because a death hasn't occurred yet, doesn't mean that it is too early to speculate about how liability should be assigned. So, let me reformulate the original question.

Consider the following hypothetical scenario. A driver tries to move away from a cyclist while passing without signaling, but the cars semi-autonomous driving systems prevent the driving from allowing sufficient space. The cyclist is clipped by the mirror, falls down and is seriously injured. How should liability be assigned? Is the driver completely liable? Or does BMW share some of the liability?

Ninety5rpm

I'll play.

The question is nonsensical. The car in question was a Level 2 AV. No way would a Level 2 prevent the driver from allowing sufficient space. There is no point about speculating about an impossible hypothetical. Same with Level 3.

Only with Level 4 is it possible that the car would override the driver's steering inputs. Realistically, I can't imagine anyone will release a Level 4 that will close-pass cyclists, so practically speaking this is probably an impossible hypothetical too.

But if something like that does happen (and there is actually a crash and injury) with a Level 4 or 5 car, certainly the manufacturer would be liable.


Reference: Levels of driving automation

I-Like-To-Bike

Got it, if YOU can't or don't imagine a specific scenario happening, than it is impossible; and as a counterpoint, if YOU do imagine a specific scenario happening, than it is a certainty and it will happen in two shakes of a lamb's tail and without any legitimate doubt of that certainty.

Ninety5rpm

Reading comprehension issues?